Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
acceptability-classification
Languages:
English
Size:
10K - 100K
ArXiv:
License:
Commit
•
0595718
1
Parent(s):
6d9e4a9
Add principle_A_domain_2 data files
Browse files- README.md +7 -3
- dataset_infos.json +6 -24
- principle_A_domain_2/train-00000-of-00001.parquet +3 -0
README.md
CHANGED
@@ -1332,10 +1332,10 @@ dataset_info:
|
|
1332 |
dtype: int32
|
1333 |
splits:
|
1334 |
- name: train
|
1335 |
-
num_bytes:
|
1336 |
num_examples: 1000
|
1337 |
-
download_size:
|
1338 |
-
dataset_size:
|
1339 |
- config_name: principle_A_domain_3
|
1340 |
features:
|
1341 |
- name: sentence_good
|
@@ -2081,6 +2081,10 @@ configs:
|
|
2081 |
data_files:
|
2082 |
- split: train
|
2083 |
path: principle_A_domain_1/train-*
|
|
|
|
|
|
|
|
|
2084 |
---
|
2085 |
|
2086 |
# Dataset Card for "blimp"
|
|
|
1332 |
dtype: int32
|
1333 |
splits:
|
1334 |
- name: train
|
1335 |
+
num_bytes: 165333
|
1336 |
num_examples: 1000
|
1337 |
+
download_size: 58464
|
1338 |
+
dataset_size: 165333
|
1339 |
- config_name: principle_A_domain_3
|
1340 |
features:
|
1341 |
- name: sentence_good
|
|
|
2081 |
data_files:
|
2082 |
- split: train
|
2083 |
path: principle_A_domain_1/train-*
|
2084 |
+
- config_name: principle_A_domain_2
|
2085 |
+
data_files:
|
2086 |
+
- split: train
|
2087 |
+
path: principle_A_domain_2/train-*
|
2088 |
---
|
2089 |
|
2090 |
# Dataset Card for "blimp"
|
dataset_infos.json
CHANGED
@@ -3135,62 +3135,50 @@
|
|
3135 |
"features": {
|
3136 |
"sentence_good": {
|
3137 |
"dtype": "string",
|
3138 |
-
"id": null,
|
3139 |
"_type": "Value"
|
3140 |
},
|
3141 |
"sentence_bad": {
|
3142 |
"dtype": "string",
|
3143 |
-
"id": null,
|
3144 |
"_type": "Value"
|
3145 |
},
|
3146 |
"field": {
|
3147 |
"dtype": "string",
|
3148 |
-
"id": null,
|
3149 |
"_type": "Value"
|
3150 |
},
|
3151 |
"linguistics_term": {
|
3152 |
"dtype": "string",
|
3153 |
-
"id": null,
|
3154 |
"_type": "Value"
|
3155 |
},
|
3156 |
"UID": {
|
3157 |
"dtype": "string",
|
3158 |
-
"id": null,
|
3159 |
"_type": "Value"
|
3160 |
},
|
3161 |
"simple_LM_method": {
|
3162 |
"dtype": "bool",
|
3163 |
-
"id": null,
|
3164 |
"_type": "Value"
|
3165 |
},
|
3166 |
"one_prefix_method": {
|
3167 |
"dtype": "bool",
|
3168 |
-
"id": null,
|
3169 |
"_type": "Value"
|
3170 |
},
|
3171 |
"two_prefix_method": {
|
3172 |
"dtype": "bool",
|
3173 |
-
"id": null,
|
3174 |
"_type": "Value"
|
3175 |
},
|
3176 |
"lexically_identical": {
|
3177 |
"dtype": "bool",
|
3178 |
-
"id": null,
|
3179 |
"_type": "Value"
|
3180 |
},
|
3181 |
"pair_id": {
|
3182 |
"dtype": "int32",
|
3183 |
-
"id": null,
|
3184 |
"_type": "Value"
|
3185 |
}
|
3186 |
},
|
3187 |
-
"supervised_keys": null,
|
3188 |
"builder_name": "blimp",
|
|
|
3189 |
"config_name": "principle_A_domain_2",
|
3190 |
"version": {
|
3191 |
"version_str": "0.1.0",
|
3192 |
-
"description": null,
|
3193 |
-
"datasets_version_to_prepare": null,
|
3194 |
"major": 0,
|
3195 |
"minor": 1,
|
3196 |
"patch": 0
|
@@ -3198,20 +3186,14 @@
|
|
3198 |
"splits": {
|
3199 |
"train": {
|
3200 |
"name": "train",
|
3201 |
-
"num_bytes":
|
3202 |
"num_examples": 1000,
|
3203 |
-
"dataset_name":
|
3204 |
-
}
|
3205 |
-
},
|
3206 |
-
"download_checksums": {
|
3207 |
-
"https://raw.githubusercontent.com/alexwarstadt/blimp/master/data/principle_A_domain_2.jsonl": {
|
3208 |
-
"num_bytes": 493189,
|
3209 |
-
"checksum": "9d14e5ae911fb8ddf571484ef2d6a6e6a053765c1c6d34253aec2c11f7d7306a"
|
3210 |
}
|
3211 |
},
|
3212 |
-
"download_size":
|
3213 |
-
"dataset_size":
|
3214 |
-
"size_in_bytes":
|
3215 |
},
|
3216 |
"principle_A_domain_3": {
|
3217 |
"description": "\nBLiMP is a challenge set for evaluating what language models (LMs) know about\nmajor grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each\ncontaining 1000 minimal pairs isolating specific contrasts in syntax,\nmorphology, or semantics. The data is automatically generated according to\nexpert-crafted grammars.\n",
|
|
|
3135 |
"features": {
|
3136 |
"sentence_good": {
|
3137 |
"dtype": "string",
|
|
|
3138 |
"_type": "Value"
|
3139 |
},
|
3140 |
"sentence_bad": {
|
3141 |
"dtype": "string",
|
|
|
3142 |
"_type": "Value"
|
3143 |
},
|
3144 |
"field": {
|
3145 |
"dtype": "string",
|
|
|
3146 |
"_type": "Value"
|
3147 |
},
|
3148 |
"linguistics_term": {
|
3149 |
"dtype": "string",
|
|
|
3150 |
"_type": "Value"
|
3151 |
},
|
3152 |
"UID": {
|
3153 |
"dtype": "string",
|
|
|
3154 |
"_type": "Value"
|
3155 |
},
|
3156 |
"simple_LM_method": {
|
3157 |
"dtype": "bool",
|
|
|
3158 |
"_type": "Value"
|
3159 |
},
|
3160 |
"one_prefix_method": {
|
3161 |
"dtype": "bool",
|
|
|
3162 |
"_type": "Value"
|
3163 |
},
|
3164 |
"two_prefix_method": {
|
3165 |
"dtype": "bool",
|
|
|
3166 |
"_type": "Value"
|
3167 |
},
|
3168 |
"lexically_identical": {
|
3169 |
"dtype": "bool",
|
|
|
3170 |
"_type": "Value"
|
3171 |
},
|
3172 |
"pair_id": {
|
3173 |
"dtype": "int32",
|
|
|
3174 |
"_type": "Value"
|
3175 |
}
|
3176 |
},
|
|
|
3177 |
"builder_name": "blimp",
|
3178 |
+
"dataset_name": "blimp",
|
3179 |
"config_name": "principle_A_domain_2",
|
3180 |
"version": {
|
3181 |
"version_str": "0.1.0",
|
|
|
|
|
3182 |
"major": 0,
|
3183 |
"minor": 1,
|
3184 |
"patch": 0
|
|
|
3186 |
"splits": {
|
3187 |
"train": {
|
3188 |
"name": "train",
|
3189 |
+
"num_bytes": 165333,
|
3190 |
"num_examples": 1000,
|
3191 |
+
"dataset_name": null
|
|
|
|
|
|
|
|
|
|
|
|
|
3192 |
}
|
3193 |
},
|
3194 |
+
"download_size": 58464,
|
3195 |
+
"dataset_size": 165333,
|
3196 |
+
"size_in_bytes": 223797
|
3197 |
},
|
3198 |
"principle_A_domain_3": {
|
3199 |
"description": "\nBLiMP is a challenge set for evaluating what language models (LMs) know about\nmajor grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each\ncontaining 1000 minimal pairs isolating specific contrasts in syntax,\nmorphology, or semantics. The data is automatically generated according to\nexpert-crafted grammars.\n",
|
principle_A_domain_2/train-00000-of-00001.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5de4982ab8ba2914d476335ba48956cc5c26e24c5ae0b64b8aed7ab06cf2903b
|
3 |
+
size 58464
|